Satellite-communications (satcom) systems operating at high power levels are subject to a unique nonlinear breakdownvoltage phenomenon in a vacuum or near-vacuum environment. Known as multi-paction effects, they can degrade the performance of a space-based system and even lead to catastrophic damage to system components. Multi-paction effects are difficult to predict but, fortunately, can be measured with a system incorporating the proper combination of hardware and software.

Such a system has been developed by In-Phase Technologies. It relies on multiple vector signal analyzers (VSAs), high-power amplification, a reliable vacuum chamber, and custom switching assemblies and software to automate and safely perform component and subsystem-level measurements of multipaction effects under realistic operating conditions. Variations of the system have been developed for applications from 1 to 18 GHz and at power levels as high as 10 kW at 2 GHz.

Multi-paction effects are generally unknown to system integrators working on terrestrial designs. They occur in vacuum or near-vacuum environments, generally in deep-space applications such as satellites, and at the relatively high power levels associated with the transmitter sections of these applications. They can affect both waveguide and coaxial components, including diplexers, filters, and antennas, as well as cables and connectors that join these components.

Multi-paction phenomena result from the activity of free electrons in a vacuum, usually excited by a high-power oscillating RF source, such as a frequency synthesizer driven by a traveling-wave-tube amplifier (TWTA), in a satellite-communications (satcom) space-based transmitter. The movement of the free electron, a charged particle, takes place in a relatively narrow gap, such as within the structure of a waveguide transmission line. In a vacuum or near-vacuum environment, the low atmospheric pressure provides conditions for oscillation for the charged particle when in proximity to a large external electric field, such as that required by the high-power TWTA in the transmitter. When the charged particle moves under the influence of the external field, it strikes the gap wall, releasing secondary charged particles. Driven into oscillation rates of millions of times per second by the external field, this process repeats and the number of free electrons multiplies, eventually leading to a multi-paction discharge event with high potential energy. Although a single discharge may cause little or no damage to the system's components, repeated multi-paction events can lead to an increase of outgassing in the system and eventually a gas discharge at a sufficient power level to cause performance degradation or critical damage to the system.

Multi-paction effects have been noted at pressures of less than 10-2 Torr. They occur in a vacuum because at atmospheric pressure, charged particles are more likely to collide with air molecules, and lose energy and velocity in the process, along with their potential to release secondary charged particles. Waveguide components designed for high-power, deep-space applications are often pressurized, to minimize the conditions necessary for multi-paction effects to occur. Multipaction effects can also be minimized by engineering components for deepspace systems with smooth edges as well as with dielectric coatings, since sharp edges within a high-frequency device can create the narrow gaps needed for discharge events at high voltage potential.

Multi-paction measurements can be made globally or locally, analyzing the performance at the system level or the component level, respectively. Global methods can determine whether a system is subject to degradation damage from multi-paction events, but they are not able to isolate the component subject to multi-paction damage in the signal-processing chain. Local measurement methods not only identify potential multi-paction weak points in a system, but are useful for monitoring portions of a system under actual operating conditions for signs of performance degradation that may occur due to multi-paction events.

Global multi-paction measurement methods tend to recreate the actual operating conditions for the entire system, including atmospheric pressure, frequency range of operation, and signal power levels. The measurements check for changes in performance, such as spikes in close-to-the-carrier phase noise, in second- and third-harmonic levels, and variations in output power, which could result from multipaction activity. Typically, such systems will sweep power levels to find a test signal power level or threshold at which performance degradation begins or damage occurs. Of course, in order to effectively evaluate a system under test for noise performance, the test system itself must produce test signals that are characterized by low phase noise, low spurious levels, and low harmonic levels. Since test signals in a multi-paction measurement system will be amplified prior to injection into the system under test, any excess test source noise can lead to false or misleading test results when evaluating the system for its multi-paction power threshold level.

Local measurement systems follow a similar approach, but are designed for testing smaller assemblies or components under vacuum or near-vacuum conditions and high-frequency, high-power levels. Both local and global measurement systems often employ modulated test signals, such as those based on amplitude modulation (AM); they can measure changes in the modulation signature from input to output of a device under test (DUT) in search of a multi-paction threshold point. Because modern satcom and other communications systems rely on advanced digital modulation formats for increased capacity per occupied signal bandwidth, an effective multi-paction measurement system must not only support digital in-phase (I) and quadrature (Q) signal modulation formats, but should also provide measurement capabilities for single- and multiple-carrier test cases.

A practical multi-paction measurement system should include the means of generating precisely controlled (and modulated) test signals, a method to amplify those signals to high power levels, and the ability to couple a safe portion of the output signal from a device or system under test to a signal analyzer. For multi-paction testing, a system under test or DUT will be contained in a vacuum chamber capable of achieving the vacuum levels of the actual application. Analysis of continuous- wave (CW) or pulsed signals may be performed by a number of instruments, including power meters, spectrum analyzers, and digital storage oscilloscopes (DSOs), although such tools as vector signal analyzers (VSAs) have proven effective in analyzing signals with digital modulation based on I/Q modulated signal formats. For any multi-paction test system, the goal is not to destroy the system under test or DUT but to find the threshold voltage and power levels at which multi-paction events begin, so that these levels can be established as absolute maximum ratings for the system or device.

One of the measurement systems developed by In-Phase Technologies for multi-paction testing is shown in Fig. 1, with a block diagram describing different elements in the system shown in Fig. 2. The system leans heavily on digitizing analog signals from a DUT by means of multiplechannel high-speed analog-to-digital converters (ADCs) for signal analysis, rather than relying on analog signal analysis by means of a microwave spectrum analyzer. The DUT remains in a vacuum chamber, with high-power coaxial transmission lines to route test signals to it and from it. The connecting cables are housed within a high-power cart (Fig. 3), which provides mobility for ease of connection to different DUTs.

The system incorporates four PXIe-5663 VSAs from National Instruments for four channels of signal generation and analysis. These are instruments-on-acard that interface to a PXI backplane capable of high-speed digital signal transfers. Each PXIe-5663 consists of a PXIe-5601 RF frequency downconverter module, a PXIe-5622 intermediate- frequency (IF) digitizer module and a PXI-5652 signal-generator module for use as a local oscillator (LO) source. The modules comprise a card-based VSA capable of measurements from 10 MHz to 6.6 GHz with sub-1-Hz tuning resolution. The highly accurate VSAs incorporate internal 10-MHz reference sources (and can operate with an external reference) with outstanding spectral purity. For the VSA itself, the typical phase noise is better than -102 dBc/ Hz offset 1 kHz from a 1-GHz carrier and better than -122 dBc/Hz offset 100 kHz from a 1-GHz carrier. The VSAs can capture signal information within 3-dB resolution bandwidths of less than 1 Hz to 10 MHz.

Continue on page 2

Page Title

In this multi-paction test system, vector signal data is continuously and synchronously acquired at a rate of 10 MSamples/s on each of the four channels. The test system uses phasenulling techniques to evaluate changes in the linearity of a component under test, comparing the quality of output signals versus that of input signals, analyzing both incident and reflected signals. Signal processing is performed in the digital realm, simplifying measurements and improving accuracy. Rather than perform manual tuning with a variable attenuator and a phase shifter to null the phase of incident and reflected test signals, this system digitizes the incident and reflected signals by means of a dual-channel, high-speed ADC and uses digitalsignal- processing (DSP) techniques to manipulate and analyze the data. Any adjustments needed between incident and reflected signals while searching for the onset of the nonlinear behavior signifying a multi-paction event can be performed quickly and accurately under digital control, rather than the more time-consuming tuning required with an attenuator and phase shifter in an analog test setup.

The ADCs in the PXIe-5663 VSAs acquire data continuously during measurements, sending measurement data to the system controller for analysis and formatting. The controller operates under custom multi-paction measurement software developed by In-Phase Technologies. The software allows a test system operator to precisely set measurement parameters, such as threshold voltages, secondand third-harmonic levels, phase nulls, amplitude levels, when searching for those conditions that can lead to the onset of a multi-paction event. The software can be programmed to triggers on various event thresholds to either store or disregard the captured signal data.

Because the multi-paction measurement system is based on digital signal analysis, it can correct for manufacturing tolerances in any passive components employed in the system in order to maintain phase-matched and phase-coherent test channels. Each measurement channel is sampled and measured for any deviations in phase and corrections are made automatically under software control. The system is calibrated for amplitude and phase variations as a function of temperature and test power levels, providing measurement results that are accurate in amplitude within 0.1 dB and in phase within 0.1 deg. across a frequency band of interest.

The multi-paction measurement system is capable of measurements at high microwave power levels, using TWTAs to achieve multiple kilowatts of microwave test power at the DUT. It performs swept power measurements, starting at low levels and ramping to a maximum programmed power level while measuring such DUT parameters as changes in return loss. Progressive or dramatic changes in return loss, for example, can be a warning that a DUT is defective or likely to suffer multi-paction discharges at high power levels. The four-channel multipaction measurement system can test as many as four DUTs sequentially, with only about 5 seconds required to switch a test sequence from one DUT to another.

Because of the complexity of multipaction measurements, test results are displayed on multiple monitors (Fig. 4) for real-time evaluation of DUT behavior under changing operating conditions. The system can also run under full software control, acquiring data continuously according to programmed requirements. As an example of an automatic measurement, the system was used to perform a device breakdown test using a 4-μs, 5-kHz pulse under swept power conditions. Measurements performed at a sample rate of 10 MSamples/s showed the appearance of a good pulse at lower power levels (Fig. 5, left), with gradual degradation in pulse characteristics until reaching shutdown 243 ms from the pretrigger point (Fig. 5, right).

A variety of these multiple-channel multi-paction measurement systems have been developed for different frequency bands from 1 to 18 GHz and for power levels as high as 10 kW at 2 GHz. The systems can perform CW or pulsed (peak) power measurements as well as measurements of DUT insertion loss, return loss, phase noise, passive intermodulation (PIM) distortion, harmonic distortion, spurious levels, and other performance parameters that can be evaluated for the onset of a multi-paction event.